I'm very optimistic for near-term AGI (10 years or less). Even just a few years ago most in the field would have said that it's an "unknown unknown", we didn't have the theory or the models, there was no path forward and so it was impossible to predict.
Now we have a fairly concrete idea of what a potential AGI might look like - an RL agent that uses a large transformer.
The issue is that unlike supervised training you need to simulate the environment along with the agent, so this requires a magnitude more compute compared to LLMs. That's why I think it will still be large corporate labs that will make the most progress in this field.
Now we have a fairly concrete idea of what a potential AGI might look like - an RL agent that uses a large transformer.
The issue is that unlike supervised training you need to simulate the environment along with the agent, so this requires a magnitude more compute compared to LLMs. That's why I think it will still be large corporate labs that will make the most progress in this field.